J . Org. Chem., Vol. 44, No. 10, 1979 1727
Notes log f H +
- log ( f X H + / f x ) = (1 - ‘$,)[log f H +
- 1%
(fBH+/fB)]
(2)
The d e parameter, depending on free-energy differences, should be a measure of the “external” stabilization of the cation, since the contributions from “internal” stabilization, almost constant through the acid range, shouid cancel out. Since positive $,’s indicate3s8J6a relatively large solvation for the cation, the decrease of its value going from acetone to benzophenone (see Table I) is a measure of the less efficient interaction between the protonated benzophenone and the solvent as a consequence of the increased “internal” charge delocalization into the aromatic rings. From the above discussion it appears that an inverse relationship should exist between “internal” and “external” stabilization of the oxycarbenium ion. Indeed, a plot of the 13C NMR chemical shift differences reported in Table I vs. the parameters gives a straight line (e.g., for A613C(FS03H),slope = -59.2, r = 0.967). Although the solvation ability for cations is a well-known property of F S O ~ H , 8 ~ 1 5no ~ 1attempt T has been made, a t the time of this writing, to separate in this solvent “internal” and “external” contributions to ion stabilization and it is therefore not possible to make a direct comparison between AH,(FSO iH) and T NMR data. It needs only to be pointed out that the relative energies of solvation for ammonium ions in FSO3H areI8 within experimental error of those in water. The fact that ~ K R Hand + M,(FSO?H)values both show the same basicity order for ketones is a further indication of the remarkably similar solvat ion properties of water and fluorosulfuric acid.18 The ‘“cand motonation data lead to the same conclusion in that’ as expected’ the positive charge is more the benzophenone than in the acetone-derived oxycarbenium ion. AS a consequence. the “internal” or intrinsic stabilization decreases in the same order. The overall stabilization of an ion, as evaluated in solution from dynamic measurements such as protonation equilibria or kinetics, is, however, a combination of “internal” and “external” (that provided by interactions with the solvent) stabilization.19 The “external” stabilization varies inversely with the “internal” one, since it is an inverse function of the charge density in the cation, and obviously depends on the nature of the solvent. I t is therefore of paramount importance, when considering dynamic measurements, to clearly identify the role played by the solvent and that by the substituents in the stabilization of charged species.
internal standard (Au = u - v,,f). Because of its low solubility in aqueous acid solutions, the NMR spectra of dicyclopropyl ketone were recorded on the Bruker WP-60 pulsed spectrometer. In this case we have monitored the peak of highest intensity in the complex multiplet of the cyclopropyl resonance. Plots of Au vs. Ho give good sigmoid curves from which the ionization ratios were obtained as described in our previous papers.3.16
Acknowledgment. One of us (G.S.) gratefully acknowledges the Alexander von Humboldt Stiftung for a fellowship and Professor Dr. W. Walter for the hospitality at the Institut fur Organische Chemie und Biochemie der Universitat, Hamburg, where part of this work was done. References and Note (1) (2) (3) (4) (5) (6) (7) (8) (9)
(10) (1 1)
(12) (13) (14)
Larsen, J. W. J. Am. Chem. SOC. 1976, 700, 330. Scorrano, G. Acc. Chem. Res. 1973, 6, 132. Levi, A.; Modena, G.; Scorrano, G. J. Am. Chem. SOC. 1974, 96,6585. Bunnett, J. F.; Olsen, F. P. Can. J. Chem. 1966, 44, 1899. Arnett, E. M.; Quirk, R. P.; Larsen. J. W. J. Am. Chem. SOC. 1970, 92, 3977. (a) McClelland, R. A.; Reynolds, W. F. Can. J. Chem. 1976, 54, 718. (b) Butler, A. R. J. Chem. SOC.,Perkin Trans. 2, 1976, 959. Olah, G. A.; Westerman, P. W.; Nishimura, J. J. Am. Chem. SOC.1974, 96, 3548. Arnett, E. M.; Scorrano, G. Adv. Phys. Org. Chem., 1976, 13, 83. The I3Cdata for the H2SOd and the FS03H solutions of the oxycarbenium ions are not identical, also because they have been computed by using different reference values for the free base. They are, however, linearly related with a slope near unity. Spiesecke, H.; Schneider, W. G. Tetrahedron Lett. 1961, 468. Karplus, M.; Pople, J. A. J. Chem. Phys. 1963, 38, 2803. Olah, G. A,; Westerman, P. W. J. Am. Chem. SOC.1973, 95,7530. Hehre, W. J.; Taft, R. W.; Tompson. R. D. Prog. Phys. Org. Chem. 1976, 72, 159; Nelson, G. L.; Williams, E. A. ibid. 1976, 72,229. Ancian, B.; Membrey. F.; Doucet, J. P. J. Org. Chem. 1978, 43, 1509. Edward, J. T.; Wong, S.C. J. Am. Chem. SOC.1977, 99,4229. Aue, D. H.; Webb, H. M.; Bowers, M. T. J. Am. Chem. SOC.1976, 98, ?,i n . Taft, R. W.: Wolf, J. F.: Beauchamo. J. L.: Scorrano. G.: Arnett. E. M. J. Am. Chem. SOC. 1978, 100, 1240. Bonvicini, P.; Levi, A,; Lucchini, V.; Mdena, G.; Scorrano, G.J. Am. Chem. SOC.1973, 95,5960. Perdoncin, G.; Scorrano. G. \bid. 1977, 99,6983. Gillespie, R. J.; Acc. Chem. Res. 1968, 1, 202. Amett, E. M., in “Proton-Transfer Reactions”, Caldin, E. F.; Gold, V.; Eds.; Chapman and Hall: London, 1975; p 79. Hepler, L. G.; O’Hara, W. F. J. Phys. Chem., 1961, 65, 81 1. Rappoport, 2. “CRC Handbook of Tables for Organic Compound Identification”, 3rd ed.; Chemical Rubber Publishing Co.; Cleveland, Ohio, 1967. Johnson, C. D.; Katritzky, A. R.; Shapiro, S. A. J. Am. Chem. SOC., 1969, 9 7, 6654. Davis, C, T,; G ~T, A, J , A ~ ~ them, . ~ sot, 1954, ~ 76,3507, ~ ~ I.,.
(151 (16) (17) (18) (19) (20) (21) (22)
1,4-Dimethyl-2,3,7-trioxabicyclo[ 2.2.11hept-5-ene: Synthesis and Characterization’
Experimental Section T h e five ketones studied were reagent grade commercial products, purified by crystallization or distillation until their physical constants agreed with accepted literature data.20 The acid solutions were made by diluting with distilled water the commercial reagent grade sulfuric acid and standardized by titration with NaOH. T h e Ho values were interpolated from published data.21The NMR spectra were recorded a t 25 “ C ( f 0 . 3 “(2) on a Bruker HFX-10 or a Bruker WP-60 spectrometers and the UV spectra on a Cary 15 spectrophotometer equipped with a thermostated cell compartment (25 f 0.2 “C). T h e solutions for the UV spectra were prepared by dissolving 0.12 g of ketone in CH2C12 (100 mL). Aliquots ( 5 mL) were then transferred in a 50-mL volumetric flask, the solvent was removed a t reduced pressure, and the residual ketone was dissolved in the thermostated acid of the appropriate concentration. The titration curves were obtained, following the Davis and Geissman method,22by plotting the differences in atlsorbance at 258 and 342 nm [AA = A258 - A342], as a function of the medium acidity, expressed by Ho. T h e ionization ratios were computed as I = ( U B - 1A)/(U - UBH+), where b i g and ~ A B Hare + the differences in absorbance for the free base and its conjugate acid, 2s obtained from the linear part of the titration curve at low and high acidity. respectively. T h e solutions for the NMR measurements were made by dissolving -20 FL of ketone in 2 mL of the sulfuric acid solutions containing Me3NH+HS04- as internal standard. The chemical shifts were measured as the difference between the methyl resonance of the methyl ketones and that of the
0022-3263/79/1944-1727$01.00/0
Waldemar Adam* and Kiyoshige Takayama Department of Chemistry, University of Puerto Rico, Rlb Piedras, Puerto Rico 00931 Received November 17, 1978
The use of 2,Ej-dimethylfuran (1) as a diagnostic test for singlet oxygen in chemical and especially biological systems is amply d ~ c u m e n t e d For . ~ example, the formation of 2,5dimethyl-2-hydroperoxy-5-hydroxydihydrofuran (3a) from 1 in the acetaldehyde-xanthine oxidase system was construed as evidence that the endoperoxide 2 intervened as the enzymatic singlet oxygenation product of 1 (eq In fact, for-
1
2
3a, R = H b. R = C H ,
01979 American Chemical Society
~
,
1728 J . Org. ('hem., Vol. 44, No. 10, 1979 mation of 3b in the photosensitized oxygenation of 1 with methanol as solcent; lent credence to this supposition. Indeed, gas-phase singlet oxygenation of 1 and condensing the product stream onto a liquid nitrogen cold finger gave the monomeric endoperoxide 2; but on warming above -30 "C it quickly polymerized with oxygen evolution.6 We now provide evidence that the monomeric endoperoxide 2 can be formed quantitat ively on photosensitized singlet oxygenation of 1 in CC14 or it is reCFC1:+a t 0-5 "C, and contrary to previous markably stable in solution even a t room temperature (ca. 25-30 "C). Our evidence includes the following: (1) a 99.3% peroxide titer (iodometry) based on 2; (2) low temperature hulb-to-bulb codistillation with CCI,; (3) conversion into the hydroperoxide 3b on treatment with methanol; and (4) quantitative deoxygenation into 1,2-diacetylethylene with t riphenylphosphine. The experimental details are described helow. Experimental Section Melting points are uncorrected. NMR spectra were run on a Hitachi Perkin-Elmer Model R-24B instrument and IR spectra on a PerkinElmer Model 2371%Infracord. Solvents and reagents used were purif'ied according to standard literature procedures. 2,.5-Dimethylfuran Endoperoxide (2). The irradiation was carried out in a 60-mL. two-neck, round-bottom flask provided with a magnetic spinhar. a rubber septum, and a gas inlet tube which was connected t o an oxygen-filled balloon. T h e flask was charged with a scrlution of' 2..5-dirnethylfuran (ca. 250 mg, 2.60 mmol) and tetraphenylporphyrin 10.~5 mg) in 20 mL of CC14 (or CFCI:d, submerged i n a water or ice bath, and positioned as close as possible (-10 cm) to a (;enera1 Electric 150-W sodium street lamp. While the solution was viyc~rouslystirred by magnetic action, the light source was activated and the required singlet oxygen was generated in situ. Within 120 min t h e p1iotooxvaenatic)n was completed, as confirmed by monitoring the progress of the reaction by NMR. The characteristic furan proton resonances (CC14. MedSi) at A 5.58 (olefinic, singlet, 2 H)and 2.17 (methyl. singlet, 6 HI were replaced by the characteristic peaks a t d 6.02 (olefinic.sing1i.t. 2 HI and 1.70 (methyl, singlet, 6 H), respectively, of the endoperoxide 2. The IR spectrum (CC14) showed bands a t 3080 (olefinic CH). 2590. 2940 (aliphatic CH). 1450, 1390, 1330, and 1210 c m - ' . Iodometric malysis of the solution gave a 95.3 f 0.5% peroxide titer based on the ,?ndoper:)xide structure 2. Attempted isolation hy s,)lvent removal e\ e n at suhamhient temperatures (-0 "C) afforded piilynieric produci, However, on flash distillation a t -20 "C (0.10 niniHg) the volatile endoperoxide codistilled intact with the CC14. 2,,j-Dimethyl-2-hydroperoxy-5-methoxy-2,5-dihydrofuran (3b). .A solution of the endoperoxide 2 (7.47 mmol) in 20 m L of CFC13 \+ascooled to -78 "C by means of a dry ice-acetone bath, and under a N:! atmosphere 6 mI, of anhydrous CH:IOHwas syringed in dropwise \\ hile the mixture pas stirred. After 2 h a t -78 "C, the reaction mixture was allowed t o warm up t o 0 " C and the solvent was removed a t 0 "(' i 10 mm Hg) to sflord 410 mg (94.5"6)of crude hydroperoxide 3b. The latter was purified by sublimation at 60-65 "C (0.15 mmHg), m p 7 1L75 "(' m p 75-76 "Ci. The 'H NMR and IR data were identical \\ith those of an authentic sample prepared directly by singlet oxygznat ion of' 2,5-dirnethylfuran in methanol. Triphenylphosphine Deoxygenation of Endoperoxide 2. A solution of the endoperoxide 2 (2.50 mmol) in 20 mL of Cc14 was cooled t o -20 "C. and while under a Na atmosphere with stirring a solution of triphenylphosiJhine (2.50 nimol) in 5 m L of CC14 was i\ringed i n at -2U " C for 1 h. The mixture was allowed to warm u p t ( i r o o i n temperature (ca -:IO " C i and was kept there for 12 h. T h e product showed t h e characteristicJa 1,2-diacetylethylene proton resonances (CC14, MeJSii at 6 6.60 (olefinic, singlet, 2 H) and 2.22 (inethyl. singlet. 6 H ) and IR (('CIA) bands a t 3050 (olefinicc-H), 1680 ( ( ' = = O )14:30, . I:i60. 1280. and 1200cm-'.
Acknowledgments. The authors would like to thank the donors of the Petroleum Research Fund, administered by the American Chemical Society (Grant No. 8341-AC-14), the National Institutes of' Health (Grant NOS. GM-22119-03, GM-00141-04. arid RR-8102-06), and the National Science Foundation (Grant No. '78-12621).
Notes References a n d Notes (1) Paper No. 80 in the "Cyclic Peroxide Series". (2) NIH Career Development Awardee, 1975-1980. (3) W. Adam, Chem.-Ztg., 99, 142 (1975). (4) (a) T. Noguchi, K . Takayama, and M. Nakano, Biochem. Biophys. Res. Commun., 78, 418 (1977). (b) E. W. Kellogg and I. Fridovich, J. Bid. Chem., 250, 8812 (1975). ( 5 ) C. S. Foote, M. T. Wuesthoff, S. Wexler, I. G. Burstain, R. Denny, G. 0 Schenck, and K. H. Schulte-Elte. Tetrahedron, 23,2583 (1967). (6) W. S.Gleason, A. D. Broadbent, E. Whittle, and J. N Pitts, Jr., J. Am. Chem. SOC.,92, 2068 (1970).
New Syntheses of Tetrathiafulvalene S. k'oneda, T. Kawase, Y. Yasuda, and Z. Yoshida* Department of S j n t h e t i c Chemistrj, K?oto C.niieratti K?oto, 606 J a p a n Receiued November 16 1978
Tetrathiafulvalene (TTF, 1) is known to be a superior 7r donor in the formation of highly conducting charge-transfer complexes.' Although this compound was first prepared by coupling of the 1,3-dithiolium cation,2 although the method by Melby, Harder, and Sheppard3 seems to be suitable for preparing large amounts of TTF, and although the procedure by Wudl et al.14 might be preferred for a small scale preparation, we report here new facile syntheses of T T F utilizing tetrakis( carbomethoxy)- (2) tetracarboxy- (3),jand dicarboxytetrathiafulvalenes (4),3,5 which have been prepared during the course of our systematic studiesfion the reactions of isotrithiones with trialkyl phosphites. A mixture of tetraester 2 and an excess of lithium bromide monohydrate in HMPA was gradually heated to 80 "C, and the temperature was maintained for 2 h (Scheme I). During this period the mixture was orange-red and gas evolution was observed. Treatment of the mixture with deaerated water gave TTF and bis(carbomethoxy)tetrathiafulvalene ( 5 ) in 11 and 53% yields, respectively.; Compound 5 was identified on the basis of its spectral data.",s Further reaction a t 150-160 "C for 10 min, after the gas evolution ceased, afforded TTF (1,13%),tetrathiafulvalenebis(N,N-dimethy1)carbamide (6, 22.3%), and tetrathiafulvalene-N,N-dimethylcarbamide ( 7 , 18.5%). The structures of 6 and 7 were determined by their analytical and spectral data. In the IH NMR spectrum of 6, the methyl protons attached to the nitrogen atom and the olefin protons appeared as two singlets a t 6 3.10 and 6.63, respectively. Generally, the methyl protons attached to the nitrogen atom of an amide group appear as a doublet because of the prominent contri,x5
Scheme I. New S y n t h e t i c Routes to P a r e n t TTF
P* r z N o C ~ ~ 6
2
Registry No.-1, 62f5-86-c5:2, 45722-89-2; 3b, 13249-74-6; 1,2tiiacetylethylene, 4 i X - 7 5 - 7 .
0022-3263/79/1944-1728$01.00/00 1979 American Chemical Society
